Structured entity querying over unstructured text

Jiang, Jiahui

Permalink

https://hdl.handle.net/2142/88060 Copy

Description

Title

Structured entity querying over unstructured text

Author(s)

Jiang, Jiahui

Issue Date

2015-07-16

Director of Research (if dissertation) or Advisor (if thesis)

Chang, Kevin Chen-chuan

Department of Study

Computer Science

Discipline

Computer Science

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

M.S.

Degree Level

Thesis

Keyword(s)

• structured query
• unstructured dataset
• rank

Abstract

The web is a collection of unstructured webpages. This characteristic makes it very hard for users to search complex queries -- in most of the times, queries are just plain text and it is very difficult for users to describe the internal relationships that they contain. The EntityRank system allows users to specify the target entities for which they are interested within the query, which brings us one step closer to being able to describe the entity relationships. But this expressiveness is still limited because the queries are still in a flat format. On the other hand, SQL queries for relational databases are very expressive. Users can easily specify multiple entities and the relationships between them. But in order to use SQL queries, we need to extract all the entities and relationships existing in the domain beforehand. The cost of maintaining the tables is very large. Also it is very hard if we want to add or modify the schema after the initial domain design. Thus we want to build a system that can combine the advantages of the flexibility and informativeness of unstructured webpages and the expressiveness of SQL queries. In this work we design a system which allows users to use structured SQL queries on unstructured webpages using the help of the EntityRank system. We design the conceptual framework to map the concepts between two systems and also propose a ranking algorithm for the final results.

Graduation Semester

2015-8

Type of Resource

text

Permalink

http://hdl.handle.net/2142/88060

Copyright and License Information

Copyright 2015 Jiahui Jiang

Owning Collections